AI Costs Surge as Vendors Shift to Usage-Based Pricing, Leaving Companies Scrambling

Major AI Vendors Abandon Flat-Rate Subscriptions

The AI industry is undergoing a fundamental pricing transformation that's sending shockwaves through corporate finance departments. In the last six months, Anthropic, OpenAI, and GitHub have shifted services away from flat-rate subscriptions toward usage-based pricing, a move that's prompting serious cost concerns among enterprise users1

1

. Microsoft has joined this trend with its premium E7 license, which bundles M365 Copilot, Agent 365, and security tools onto E51

1

. GitHub moved its Copilot plans to usage-based billing in June, while OpenAI added pay-as-you-go Codex seats in April2

2

. Anthropic removed Claude 5 from its standard subscriptions and seat-based models over difficult-to-predict demand2

2

.

Software Budgets Expected to Soar Amid AI-Driven Increases

Forrester research, based on a survey of more than 2,600 business and technology decision-makers, reveals that software budgets are expected to rise as vendors increase prices or add usage charges to pass their AI costs to customers1

1

. The shift from flat rates to usage-based fees introduces multiple variables including model selection, context size, output length, and agent operating time, leading to far more unpredictable outgoings2

2

. More than four in five leaders expect to increase overall budgets over the next 12 months, with 82% of tech decision-makers expecting larger budgets2

2

. Forrester found that 80 percent of decision-makers expect data and AI spending budgets to rise1

1

. Last year, consultants Bain & Company estimated that the build cost for AI datacenters would hit $2 trillion by 20301

1

.

Consumption-Based Pricing Models Create Unexpected Financial Challenges

The transition to consumption-based pricing models has exposed a critical gap in corporate financial management. KPMG research found that nearly a third of corporate leaders reported difficulty understanding and controlling operating costs when implementing business AI at scale1

1

. The accounting firm spoke with 2,145 executives around the world, and one-third said they had a limited understanding of usage costs4

4

. Rob Fisher, global head of advisory at KPMG, stated: "AI is now as much a financial management priority as it is a technology one. The real risk isn't investing in AI but doing so without cost visibility and an understanding of the economics of AI"4

4

. As agentic AI reshapes the enterprise Software-as-a-Service landscape, the friction between AI innovation and fiscal predictability is fast approaching a stalemate in software procurement3

3

.

Token Consumption Drives Unpredictability in Costs

With AI pricing models now tied to token consumption, every interaction with a model consumes tokens—the small units of text it reads and writes—and organizations pay for each5

5

. The meter is always running, and most organizations cannot see it move until it is too late to do anything about it5

5

. Treating AI cost as something reconciled after the fact is how good products become unprofitable ones5

5

. The confusing part is that AI keeps getting cheaper to use while the bills keep climbing. Gartner forecasts that by 2030, running inference costs on a one-trillion-parameter model will cost providers more than 90% less than it did in 20255

5

. Yet enterprise AI spending is rising anyway, because consumption is growing faster than prices are dropping5

5

.

Managing AI Costs Requires New FinOps Capabilities

Forrester recommends that organizations adapt their FinOps practices to help manage the unpredictable costs associated with AI. "Traditional FinOps wasn't built for token-based, usage-driven AI costs, but that team is certainly best positioned to build these new capabilities and must make this leap in 2027," the report stated1

1

. The research firm recommends funding runtime cost controls such as model routing, semantic caching, and usage guardrails to prevent runaway spend1

1

. Sharyn Leaver, chief research officer at Forrester, emphasized: "The organizations that outperform in 2027 won't be those that spend the most on AI. They'll be the ones that invest in the foundations that make AI effective: trusted data, strong governance, organizational readiness, and the ability to continuously adapt as technology and customer behavior evolve"1

1

. Organizations that have cost visibility and maintain strong oversight are the ones translating AI investment into real, measurable value4

4

.